X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=include%2Fsocgraph.php;h=69d33089869dbc7b8acd574904d81b39fd1651e0;hb=a770be3cb6fdc305712da58e188de4e82f3b7234;hp=272fcea9945d955764f11b9d4efa9a588001c6bf;hpb=201d95a8558506ede370e3f4d1986428fd6826bb;p=friendica.git diff --git a/include/socgraph.php b/include/socgraph.php index 272fcea994..69d3308986 100644 --- a/include/socgraph.php +++ b/include/socgraph.php @@ -7,6 +7,8 @@ require_once("include/html2bbcode.php"); /* To-Do: - Move GNU Social URL schemata (http://server.tld/user/number) to http://server.tld/username + - Fetch profile data from profile page for Redmatrix users + - Detect if it is a forum */ /* @@ -169,6 +171,8 @@ function poco_check($profile_url, $name, $network, $profile_photo, $about, $loca $gcid = ""; + $alternate = poco_alternate_ostatus_url($profile_url); + if ($profile_url == "") return $gcid; @@ -176,7 +180,7 @@ function poco_check($profile_url, $name, $network, $profile_photo, $about, $loca if (!isset($urlparts["scheme"])) return $gcid; - if (in_array($urlparts["host"], array("facebook.com", "twitter.com", "www.twitter-rss.com", + if (in_array($urlparts["host"], array("www.facebook.com", "facebook.com", "twitter.com", "identi.ca", "alpha.app.net"))) return $gcid; @@ -203,7 +207,7 @@ function poco_check($profile_url, $name, $network, $profile_photo, $about, $loca ); if(count($r)) { $network = $r[0]["network"]; - $profile_url = $r[0]["url"]; + //$profile_url = $r[0]["url"]; } } @@ -219,31 +223,48 @@ function poco_check($profile_url, $name, $network, $profile_photo, $about, $loca $updated = $x[0]["updated"]; $created = $x[0]["created"]; - $last_contact = $x[0]["last_contact"]; - $last_failure = $x[0]["last_failure"]; $server_url = $x[0]["server_url"]; $nick = $x[0]["nick"]; + $addr = $x[0]["addr"]; } else { $created = "0000-00-00 00:00:00"; - $last_contact = "0000-00-00 00:00:00"; - $last_failure = "0000-00-00 00:00:00"; $server_url = ""; $urlparts = parse_url($profile_url); $nick = end(explode("/", $urlparts["path"])); + $addr = ""; } - if ((($network == "") OR ($name == "") OR ($profile_photo == "") OR ($server_url == "")) - AND poco_reachable($profile_url, $server_url, $network)) { + if ((($network == "") OR ($name == "") OR ($addr == "") OR ($profile_photo == "") OR ($server_url == "") OR $alternate) + AND poco_reachable($profile_url, $server_url, $network, false)) { $data = probe_url($profile_url); + + $orig_profile = $profile_url; + $network = $data["network"]; $name = $data["name"]; $nick = $data["nick"]; + $addr = $data["addr"]; $profile_url = $data["url"]; $profile_photo = $data["photo"]; $server_url = $data["baseurl"]; + + if ($alternate AND ($network == NETWORK_OSTATUS)) { + // Delete the old entry - if it exists + $r = q("SELECT `id` FROM `gcontact` WHERE `nurl` = '%s'", dbesc(normalise_link($orig_profile))); + if ($r) { + q("DELETE FROM `gcontact` WHERE `nurl` = '%s'", dbesc(normalise_link($orig_profile))); + q("DELETE FROM `glink` WHERE `gcid` = %d", intval($r[0]["id"])); + } + + // possibly create a new entry + poco_check($profile_url, $name, $network, $profile_photo, $about, $location, $gender, $keywords, $connect_url, $updated, $generation, $cid, $uid, $zcid); + } } + if ($alternate AND ($network == NETWORK_OSTATUS)) + return $gcid; + if (count($x) AND ($x[0]["network"] == "") AND ($network != "")) { q("UPDATE `gcontact` SET `network` = '%s' WHERE `nurl` = '%s'", dbesc($network), @@ -261,19 +282,6 @@ function poco_check($profile_url, $name, $network, $profile_photo, $about, $loca poco_check_server($server_url, $network); - // Only fetch last update manually if it wasn't provided and enabled in the system - if (get_config('system','poco_completion') AND ($orig_updated == "0000-00-00 00:00:00") - AND poco_do_update($created, $updated, $last_failure, $last_contact) - AND poco_reachable($profile_url, $server_url, $network)) { - $last_updated = poco_last_updated($profile_url); - if ($last_updated) { - $updated = $last_updated; - $last_contact = datetime_convert(); - logger("Last updated for profile ".$profile_url.": ".$updated, LOGGER_DEBUG); - } else - $last_failure = datetime_convert(); - } - if(count($x)) { $gcid = $x[0]['id']; @@ -289,14 +297,18 @@ function poco_check($profile_url, $name, $network, $profile_photo, $about, $loca if (($keywords == "") AND ($x[0]['keywords'] != "")) $keywords = $x[0]['keywords']; + if (($addr == "") AND ($x[0]['addr'] != "")) + $addr = $x[0]['addr']; + if (($generation == 0) AND ($x[0]['generation'] > 0)) $generation = $x[0]['generation']; if($x[0]['name'] != $name || $x[0]['photo'] != $profile_photo || $x[0]['updated'] < $updated) { - q("UPDATE `gcontact` SET `name` = '%s', `network` = '%s', `photo` = '%s', `connect` = '%s', `url` = '%s', `server_url` = '%s', + q("UPDATE `gcontact` SET `name` = '%s', `addr` = '%s', `network` = '%s', `photo` = '%s', `connect` = '%s', `url` = '%s', `server_url` = '%s', `updated` = '%s', `location` = '%s', `about` = '%s', `keywords` = '%s', `gender` = '%s', `generation` = %d WHERE (`generation` >= %d OR `generation` = 0) AND `nurl` = '%s'", dbesc($name), + dbesc($addr), dbesc($network), dbesc($profile_photo), dbesc($connect_url), @@ -313,10 +325,11 @@ function poco_check($profile_url, $name, $network, $profile_photo, $about, $loca ); } } else { - q("INSERT INTO `gcontact` (`name`, `nick`, `network`, `url`, `nurl`, `photo`, `connect`, `server_url`, `created`, `updated`, `last_contact`, `last_failure`, `location`, `about`, `keywords`, `gender`, `generation`) - VALUES ('%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', %d)", + q("INSERT INTO `gcontact` (`name`, `nick`, `addr`, `network`, `url`, `nurl`, `photo`, `connect`, `server_url`, `created`, `updated`, `location`, `about`, `keywords`, `gender`, `generation`) + VALUES ('%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', %d)", dbesc($name), dbesc($nick), + dbesc($addr), dbesc($network), dbesc($profile_url), dbesc(normalise_link($profile_url)), @@ -325,8 +338,6 @@ function poco_check($profile_url, $name, $network, $profile_photo, $about, $loca dbesc($server_url), dbesc(datetime_convert()), dbesc($updated), - dbesc($last_contact), - dbesc($last_failure), dbesc($location), dbesc($about), dbesc($keywords), @@ -377,7 +388,7 @@ function poco_check($profile_url, $name, $network, $profile_photo, $about, $loca return $gcid; } -function poco_reachable($profile, $server = "", $network = "") { +function poco_reachable($profile, $server = "", $network = "", $force = false) { if ($server == "") $server = poco_detect_server($profile); @@ -385,7 +396,7 @@ function poco_reachable($profile, $server = "", $network = "") { if ($server == "") return true; - return poco_check_server($server, $network); + return poco_check_server($server, $network, $force); } function poco_detect_server($profile) { @@ -420,7 +431,11 @@ function poco_detect_server($profile) { return $server_url; } -function poco_last_updated($profile) { +function poco_alternate_ostatus_url($url) { + return(preg_match("=https?://.+/user/\d+=ism", $url, $matches)); +} + +function poco_last_updated($profile, $force = false) { $gcontacts = q("SELECT * FROM `gcontact` WHERE `nurl` = '%s'", dbesc(normalise_link($profile))); @@ -434,9 +449,30 @@ function poco_last_updated($profile) { else $server_url = poco_detect_server($profile); - if ($server_url != "") - if (!poco_check_server($pserver_url, $gcontacts[0]["network"])) + if ($server_url != "") { + if (!poco_check_server($server_url, $gcontacts[0]["network"], $force)) { + + if ($force) + q("UPDATE `gcontact` SET `last_failure` = '%s' WHERE `nurl` = '%s'", + dbesc(datetime_convert()), dbesc(normalise_link($profile))); + return false; + } + + q("UPDATE `gcontact` SET `server_url` = '%s' WHERE `nurl` = '%s'", + dbesc($server_url), dbesc(normalise_link($profile))); + } + + if (in_array($gcontacts[0]["network"], array("", NETWORK_FEED))) { + $server = q("SELECT `network` FROM `gserver` WHERE `nurl` = '%s' AND `network` != ''", + dbesc(normalise_link($server_url))); + + if ($server) + q("UPDATE `gcontact` SET `network` = '%s' WHERE `nurl` = '%s'", + dbesc($server[0]["network"]), dbesc(normalise_link($profile))); + else + return; + } // noscrape is really fast so we don't cache the call. if (($gcontacts[0]["server_url"] != "") AND ($gcontacts[0]["nick"] != "")) { @@ -446,12 +482,14 @@ function poco_last_updated($profile) { if ($server) { $noscraperet = z_fetch_url($server[0]["noscrape"]."/".$gcontacts[0]["nick"]); - if ($noscraperet["success"]) { + + if ($noscraperet["success"] AND ($noscraperet["body"] != "")) { + $noscrape = json_decode($noscraperet["body"], true); - if (($noscrape["name"] != "") AND ($noscrape["name"] != $gcontacts[0]["name"])) + if (($noscrape["fn"] != "") AND ($noscrape["fn"] != $gcontacts[0]["name"])) q("UPDATE `gcontact` SET `name` = '%s' WHERE `nurl` = '%s'", - dbesc($noscrape["name"]), dbesc(normalise_link($profile))); + dbesc($noscrape["fn"]), dbesc(normalise_link($profile))); if (($noscrape["photo"] != "") AND ($noscrape["photo"] != $gcontacts[0]["photo"])) q("UPDATE `gcontact` SET `photo` = '%s' WHERE `nurl` = '%s'", @@ -465,10 +503,18 @@ function poco_last_updated($profile) { q("UPDATE `gcontact` SET `gender` = '%s' WHERE `nurl` = '%s'", dbesc($noscrape["gender"]), dbesc(normalise_link($profile))); - if (($noscrape["about"] != "") AND ($noscrape["about"] != $gcontacts[0]["name"])) + if (($noscrape["pdesc"] != "") AND ($noscrape["pdesc"] != $gcontacts[0]["about"])) + q("UPDATE `gcontact` SET `about` = '%s' WHERE `nurl` = '%s'", + dbesc($noscrape["pdesc"]), dbesc(normalise_link($profile))); + + if (($noscrape["about"] != "") AND ($noscrape["about"] != $gcontacts[0]["about"])) q("UPDATE `gcontact` SET `about` = '%s' WHERE `nurl` = '%s'", dbesc($noscrape["about"]), dbesc(normalise_link($profile))); + if (isset($noscrape["comm"]) AND ($noscrape["comm"] != $gcontacts[0]["community"])) + q("UPDATE `gcontact` SET `community` = %d WHERE `nurl` = '%s'", + intval($noscrape["comm"]), dbesc(normalise_link($profile))); + if (isset($noscrape["tags"])) $keywords = implode(" ", $noscrape["tags"]); else @@ -509,12 +555,30 @@ function poco_last_updated($profile) { } // If we only can poll the feed, then we only do this once a while - if (!poco_do_update($gcontacts[0]["created"], $gcontacts[0]["updated"], $gcontacts[0]["last_failure"], $gcontacts[0]["last_contact"])) + if (!$force AND !poco_do_update($gcontacts[0]["created"], $gcontacts[0]["updated"], $gcontacts[0]["last_failure"], $gcontacts[0]["last_contact"])) return $gcontacts[0]["updated"]; $data = probe_url($profile); - if (($data["poll"] == "") OR ($data["network"] == NETWORK_FEED)) { + // Is the profile link the alternate OStatus link notation? (http://domain.tld/user/4711) + // Then check the other link and delete this one + if (($data["network"] == NETWORK_OSTATUS) AND poco_alternate_ostatus_url($profile) AND + (normalise_link($profile) == normalise_link($data["alias"])) AND + (normalise_link($profile) != normalise_link($data["url"]))) { + + // Delete the old entry + q("DELETE FROM `gcontact` WHERE `nurl` = '%s'", dbesc(normalise_link($profile))); + q("DELETE FROM `glink` WHERE `gcid` = %d", intval($gcontacts[0]["id"])); + + poco_check($data["url"], $data["name"], $data["network"], $data["photo"], $gcontacts[0]["about"], $gcontacts[0]["location"], + $gcontacts[0]["gender"], $gcontacts[0]["keywords"], $data["addr"], $gcontacts[0]["updated"], $gcontacts[0]["generation"]); + + poco_last_updated($data["url"], $force); + + return false; + } + + if (($data["poll"] == "") OR (in_array($data["network"], array(NETWORK_FEED, NETWORK_PHANTOM)))) { q("UPDATE `gcontact` SET `last_failure` = '%s' WHERE `nurl` = '%s'", dbesc(datetime_convert()), dbesc(normalise_link($profile))); return false; @@ -608,8 +672,8 @@ function poco_do_update($created, $updated, $last_failure, $last_contact) { return false; // If the last contact was less than a week ago and the last failure is older than a week then don't update - if ((($now - $contact_time) < (60 * 60 * 24 * 7)) AND ($contact_time > $failure_time)) - return false; + //if ((($now - $contact_time) < (60 * 60 * 24 * 7)) AND ($contact_time > $failure_time)) + // return false; // If the last contact time was more than a week ago and the contact was created more than a week ago, then only try once a week if ((($now - $contact_time) > (60 * 60 * 24 * 7)) AND (($now - $created_time) > (60 * 60 * 24 * 7)) AND (($now - $failure_time) < (60 * 60 * 24 * 7))) @@ -673,7 +737,7 @@ function poco_check_server($server_url, $network = "", $force = false) { $last_contact = "0000-00-00 00:00:00"; $last_failure = "0000-00-00 00:00:00"; } - logger("Server ".$server_url." is unknown. Start discovery.", LOGGER_DEBUG); + logger("Server ".$server_url." is outdated or unknown. Start discovery. Force: ".$force." Created: ".$servers[0]["created"]." Failure: ".$last_failure." Contact: ".$last_contact, LOGGER_DEBUG); $failure = false; $orig_last_failure = $last_failure; @@ -692,8 +756,11 @@ function poco_check_server($server_url, $network = "", $force = false) { } if (!$serverret["success"] OR ($serverret["body"] == "") OR (sizeof($xmlobj) == 0) OR !is_object($xmlobj)) { - $last_failure = datetime_convert(); - $failure = true; + // Workaround for bad configured servers (known nginx problem) + if ($serverret["debug"]["http_code"] != "403") { + $last_failure = datetime_convert(); + $failure = true; + } } elseif ($network == NETWORK_DIASPORA) $last_contact = datetime_convert(); @@ -980,15 +1047,16 @@ function count_common_friends($uid,$cid) { $r = q("SELECT count(*) as `total` FROM `glink` INNER JOIN `gcontact` on `glink`.`gcid` = `gcontact`.`id` - where `glink`.`cid` = %d and `glink`.`uid` = %d - and `gcontact`.`nurl` in (select nurl from contact where uid = %d and self = 0 and blocked = 0 and hidden = 0 and id != %d ) ", + WHERE `glink`.`cid` = %d AND `glink`.`uid` = %d AND + ((`gcontact`.`last_contact` >= `gcontact`.`last_failure`) OR (`gcontact`.`updated` >= `gcontact`.`last_failure`)) + AND `gcontact`.`nurl` IN (select nurl from contact where uid = %d and self = 0 and blocked = 0 and hidden = 0 and id != %d ) ", intval($cid), intval($uid), intval($uid), intval($cid) ); -// logger("count_common_friends: $uid $cid {$r[0]['total']}"); +// logger("count_common_friends: $uid $cid {$r[0]['total']}"); if(count($r)) return $r[0]['total']; return 0; @@ -1003,11 +1071,15 @@ function common_friends($uid,$cid,$start = 0,$limit=9999,$shuffle = false) { else $sql_extra = " order by `gcontact`.`name` asc "; - $r = q("SELECT `gcontact`.* - FROM `glink` INNER JOIN `gcontact` on `glink`.`gcid` = `gcontact`.`id` - where `glink`.`cid` = %d and `glink`.`uid` = %d - and `gcontact`.`nurl` in (select nurl from contact where uid = %d and self = 0 and blocked = 0 and hidden = 0 and id != %d ) - $sql_extra limit %d, %d", + $r = q("SELECT `gcontact`.*, `contact`.`id` AS `cid` + FROM `glink` + INNER JOIN `gcontact` ON `glink`.`gcid` = `gcontact`.`id` + INNER JOIN `contact` ON `gcontact`.`nurl` = `contact`.`nurl` + WHERE `glink`.`cid` = %d and `glink`.`uid` = %d + AND `contact`.`uid` = %d AND `contact`.`self` = 0 AND `contact`.`blocked` = 0 + AND `contact`.`hidden` = 0 AND `contact`.`id` != %d + AND ((`gcontact`.`last_contact` >= `gcontact`.`last_failure`) OR (`gcontact`.`updated` >= `gcontact`.`last_failure`)) + $sql_extra LIMIT %d, %d", intval($cid), intval($uid), intval($uid), @@ -1064,7 +1136,8 @@ function count_all_friends($uid,$cid) { $r = q("SELECT count(*) as `total` FROM `glink` INNER JOIN `gcontact` on `glink`.`gcid` = `gcontact`.`id` - where `glink`.`cid` = %d and `glink`.`uid` = %d ", + where `glink`.`cid` = %d and `glink`.`uid` = %d AND + ((`gcontact`.`last_contact` >= `gcontact`.`last_failure`) OR (`gcontact`.`updated` >= `gcontact`.`last_failure`))", intval($cid), intval($uid) ); @@ -1078,10 +1151,14 @@ function count_all_friends($uid,$cid) { function all_friends($uid,$cid,$start = 0, $limit = 80) { - $r = q("SELECT `gcontact`.* - FROM `glink` INNER JOIN `gcontact` on `glink`.`gcid` = `gcontact`.`id` - where `glink`.`cid` = %d and `glink`.`uid` = %d - order by `gcontact`.`name` asc LIMIT %d, %d ", + $r = q("SELECT `gcontact`.*, `contact`.`id` AS `cid` + FROM `glink` + INNER JOIN `gcontact` on `glink`.`gcid` = `gcontact`.`id` + LEFT JOIN `contact` ON `contact`.`nurl` = `gcontact`.`nurl` AND `contact`.`uid` = %d + WHERE `glink`.`cid` = %d AND `glink`.`uid` = %d AND + ((`gcontact`.`last_contact` >= `gcontact`.`last_failure`) OR (`gcontact`.`updated` >= `gcontact`.`last_failure`)) + ORDER BY `gcontact`.`name` ASC LIMIT %d, %d ", + intval($uid), intval($cid), intval($uid), intval($start), @@ -1168,8 +1245,8 @@ function update_suggestions() { $done[] = $a->get_baseurl() . '/poco'; - if(strlen(get_config('system','directory_submit_url'))) { - $x = fetch_url('http://dir.friendica.com/pubsites'); + if(strlen(get_config('system','directory'))) { + $x = fetch_url(get_server()."/pubsites"); if($x) { $j = json_decode($x); if($j->entries) { @@ -1199,18 +1276,52 @@ function update_suggestions() { } } +function poco_discover_federation() { + $last = get_config('poco','last_federation_discovery'); + + if($last) { + $next = $last + (24 * 60 * 60); + if($next > time()) + return; + } + + $serverdata = fetch_url("http://the-federation.info/pods.json"); + + if (!$serverdata) + return; + + $servers = json_decode($serverdata); + + foreach($servers->pods AS $server) + poco_check_server("https://".$server->host); + + set_config('poco','last_federation_discovery', time()); + +} + function poco_discover($complete = false) { + // Update the server list + poco_discover_federation(); + $no_of_queries = 5; - $last_update = date("c", time() - (60 * 60 * 6)); // 24 + $requery_days = intval(get_config("system", "poco_requery_days")); + + if ($requery_days == 0) + $requery_days = 7; - $r = q("SELECT `poco`, `nurl`, `url`, `network` FROM `gserver` WHERE `last_contact` > `last_failure` AND `poco` != '' AND `last_poco_query` < '%s' ORDER BY RAND()", dbesc($last_update)); + $last_update = date("c", time() - (60 * 60 * 24 * $requery_days)); + + $r = q("SELECT `poco`, `nurl`, `url`, `network` FROM `gserver` WHERE `last_contact` >= `last_failure` AND `poco` != '' AND `last_poco_query` < '%s' ORDER BY RAND()", dbesc($last_update)); if ($r) foreach ($r AS $server) { - if (!poco_check_server($server["url"], $server["network"])) + if (!poco_check_server($server["url"], $server["network"])) { + // The server is not reachable? Okay, then we will try it later + q("UPDATE `gserver` SET `last_poco_query` = '%s' WHERE `nurl` = '%s'", dbesc(datetime_convert()), dbesc($server["nurl"])); continue; + } // Fetch all users from the other server $url = $server["poco"]."/?fields=displayName,urls,photos,updated,network,aboutMe,currentLocation,tags,gender,generation"; @@ -1251,8 +1362,13 @@ function poco_discover($complete = false) { q("UPDATE `gserver` SET `last_poco_query` = '%s' WHERE `nurl` = '%s'", dbesc(datetime_convert()), dbesc($server["nurl"])); if (!$complete AND (--$no_of_queries == 0)) break; - } else // If the server hadn't replied correctly, then force a sanity check + } else { + // If the server hadn't replied correctly, then force a sanity check poco_check_server($server["url"], $server["network"], true); + + // If we couldn't reach the server, we will try it some time later + q("UPDATE `gserver` SET `last_poco_query` = '%s' WHERE `nurl` = '%s'", dbesc(datetime_convert()), dbesc($server["nurl"])); + } } }